He states that the relative returns of different styles do not persist, but that their relative risks (as measured by volatility) do persist. He presents evidence that risk adjusted returns (as measured by the Sharpe ratio) depend heavily upon investment styles.
The net effect is that costs always dominate all other factors, but that investment styles cause differences in risk adjusted returns.
About Risk Adjusted Returns
John Bogle makes an important observation in the box on page 150.
I will add a subtle point: withdrawals during retirement (and deposits during accumulation) influence the actual holding period. In fact, making withdrawals means having many holding periods.Further, weighting risk as equal to return in importance in the formula is completely arbitrary. Here is the reality of investing, as I see it: An extra percentage point of standard deviation is meaningless, but an extra percentage point of return is priceless.
The effects of volatility diminish with time while the effect relative returns persist. If you are choosing between two investments and your holding period is long, focus on their relative returns.
Expanding the Risk Adjusted Return Discussion
There is an important set of theorems based upon the ratio of return to risk (in this case, risk is taken to be the standard deviation, not the variance). If you are optimizing a portfolio for the greatest return for a specified amount of risk, there will be a single, best combination of investments. You should leverage that combination until it has the amount of risk that you are willing to accept. [This is a first approximation. It ignores a variety of details that would argue against ever using leverage.] The final combination of investments does not change. The amount of leverage does.
That is the starting point, but not the ending point. Each time, as you throw in additional distinct factors for optimization, you pick up another dimension. That is, when you add a third factor for optimization, there is still a unique combination of investments that you should leverage, but it may be different from the original combination. Having the third factor may exclude selecting the original combination. When you throw in a fourth factor, there may be a different, unique combination. Once again, when you add the new factor, you may exclude a previously determined, optimal combination of investments. And so forth.
If we have different investors, each with at least three factors to optimize, and if their factors (other than the risk-adjusted return) differ, you are highly likely to end up with multiple, different optimal combinations of investments. Their best choices differ because those other factors differ.
The original theorem is excellent as a starting point. It is demonstrably wrong if pushed to an extreme. That is an error that many people make.
John Bogle introduced a third factor: a long-term outlook. Because of his long-term perspective, mutual fund costs (which persist and which include more than fees) dominate many other investment decisions.
John Bogle versus William Bernstein
I attach much more credibility to John Bogle's writings that to those found in William Bernstein's The Four Pillars of Investing. The reason is John Bogle identifies the sources of his numbers and I can find out when he makes a minor misstatement (such as referring to average returns instead of annualized return). He is very up front in presenting his evidence.
William Bernstein provides a sharp contrast. He presents many plausibility arguments that generally support his numbers, but not the sources that led to his numbers.
I offer as an example his words on page 234 of The Four Pillars of Investing.
I have never been able to determine exactly where his 2% number came from. Why not 1.5%?In other words, a particular bad returns sequence can reduce your safe withdrawal amount by as much as 2% below the long-term return of stocks. Recall from Chapter 2 that it's likely that future stock returns will be in the 3.5% range, which means that current retirees may not be entirely safe withdrawing more than 2% of the real starting values of their portfolios per year!
My best guess, based on his comments at the bottom of page 234 and the top of page 235, is that his 2% number came from a Monte Carlo model. His discussions (and numbers) associated with the sequence of returns were nothing more than plausibility arguments.
An alternative hypothesis is that William Bernstein did not like his answer. He had calculated 1.5% and he did not feel comfortable defending 1.5%. He hedged.
Regardless of the reason, William Bernstein is hiding something from us. John Bogle is always up front.
Have fun.
John R.